Internet of vehicles connection management platform management and control processing method, device, equipment and medium

Abstract

The invention discloses an Internet of Vehicles connection management platform management and control processing method, device, equipment and medium, and the method comprises the steps: collecting real-time data of an Internet of Vehicles connection management platform, the real-time data comprising order system data, system monitoring data and package data; preprocessing the collected real-time data of the Internet of Vehicles connection management platform to obtain preprocessed data of the Internet of Vehicles connection management platform; according to the preprocessed data, time sequence embedding features are extracted through a long-short-term memory network model with a time attention mechanism, and in combination with service features and system monitoring data in the preprocessed data, an order blockage risk probability is obtained through a limit gradient improvement model; and calculating an order system load factor according to system monitoring data, calculating a conventional process influence degree in combination with a management and control order proportion and a time sensitive factor, and judging whether to trigger a management and control short process according to the conventional process influence degree and an order blockage risk probability.

Description

Technical Field

[0001] This invention relates to the field of Internet of Things (IoT) management technology, and in particular to a method, apparatus, equipment, and medium for managing and controlling a vehicle network connection management platform. Background Technology

[0002] Existing Vehicle Connection Management Platforms (VCMPs) are prone to high concurrency and order congestion during peak periods of bulk business. The current billing and control process requires a complete closed loop involving the billing system, order generation, order system queuing, and service activation execution. This results in controlled orders (such as shutdowns and network outages) needing to be processed in queues, leading to delays in issuing control instructions. This can cause issues such as "users are in arrears but service is not interrupted in time," resulting in resource waste and potentially impacting business compliance due to untimely service control. Current optimization methods for high order concurrency typically focus on improving the order system's own processing capacity (such as queue expansion and database sharding), lacking differentiated processing mechanisms for high-priority controlled orders. Summary of the Invention

[0003] The technical problem to be solved by the present invention is to provide a vehicle network connection management platform control and processing method, device, equipment and medium to solve the technical problems of resource waste caused by delayed issuance of control commands and lack of priority processing of key business.

[0004] Firstly, a method for managing and controlling vehicle-to-everything (V2X) connectivity platforms is provided, including the following steps: Collect real-time data from the vehicle network connection management platform, including order system data, system monitoring data, and package data; The real-time data collected from the vehicle network connection management platform is preprocessed to obtain preprocessed data from the vehicle network connection management platform. Based on the preprocessed data, temporal embedding features are extracted using a long short-term memory network model with a time attention mechanism. Combined with business features and system monitoring data in the preprocessed data, the probability of order blockage risk is obtained using a limit gradient boosting model. The order system load factor is calculated based on system monitoring data. The impact of the regular process is calculated by combining the proportion of controlled orders and the time-sensitive factor. Based on the impact of the regular process and the probability of order congestion, it is determined whether to trigger the controlled short process.

[0005] Secondly, a vehicle network connectivity management platform control and processing device is provided, including: The data acquisition module is used to collect real-time data from the vehicle network connection management platform, wherein the real-time data includes order system data, system monitoring data, and package data; The data preprocessing module is used to preprocess the real-time data collected from the vehicle network connection management platform to obtain preprocessed data from the vehicle network connection management platform. The fusion prediction module is used to extract temporal embedding features from the preprocessed data using a long short-term memory network model with a time attention mechanism, and to obtain the probability of order congestion risk by combining business features and system monitoring data in the preprocessed data with a limit gradient boosting model. The dynamic decision-making module is used to calculate the order system load factor based on system monitoring data, combine the proportion of controlled orders and time-sensitive factors to calculate the impact of regular processes, and determine whether to trigger a controlled short process based on the impact of regular processes and the probability of order congestion risk.

[0006] Thirdly, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the above-mentioned vehicle network connection management platform control and processing method.

[0007] Fourthly, a computer-readable storage medium is provided, which stores a computer program that, when executed by a processor, implements the steps of the above-mentioned vehicle network connection management platform control and processing method.

[0008] The solution implemented by the aforementioned vehicle network connectivity management platform control and processing method, device, equipment, and medium can collect real-time data from the vehicle network connectivity management platform, preprocess it, and then use a long short-term memory network model with a time attention mechanism combined with a limit gradient boosting model to predict the probability of order congestion. This fully utilizes multi-dimensional information and is accurate and reliable, enabling proactive prediction of order congestion risks. This solves the problem of resource waste caused by the backlog and delay in issuing control instructions due to unpredictable order congestion. Based on the probability of order congestion risk and the impact of regular processes, the system can determine the triggering of short control processes. This ensures that most regular orders are processed normally while intelligently identifying and triggering short processes for high-priority control orders, avoiding excessive impact on system data consistency due to the abuse of short processes, and ensuring the priority processing of core control business. Attached Figure Description

[0009] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0010] Figure 1 This is a flowchart illustrating a vehicle network connection management platform control and processing method according to an embodiment of the present invention; Figure 2 for Figure 1 A flowchart illustrating a specific implementation of step S20; Figure 3 for Figure 1 A flowchart illustrating a specific implementation of step S30; Figure 4 This is a schematic diagram of the structure of a vehicle network connection management platform control and processing device according to an embodiment of the present invention; Figure 5 This is a schematic diagram of the structure of a computer device according to an embodiment of the present invention; Figure 6 This is another structural schematic diagram of a computer device according to one embodiment of the present invention. Detailed Implementation

[0011] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0012] Please see Figure 1 As shown, Figure 1 A flowchart illustrating the vehicle network connectivity management platform control and processing method provided in this embodiment of the invention includes the following steps: S10: Collect real-time data from the vehicle network connection management platform. This real-time data includes order system data, system monitoring data, and package data. The order system data includes real-time order volume, order growth rate, length of the pending order message queue, and average order processing time. The system monitoring data includes CPU (Central Processing Unit) utilization, memory utilization, and database read / write latency. The package data includes the number of expired packages, the number of data-overloaded SIM cards, and the number of renewed packages. The order growth rate is the growth rate of orders within a preset time interval, which can be 5 minutes.

[0013] S20: Preprocess the real-time data collected from the vehicle network connection management platform to obtain preprocessed data from the vehicle network connection management platform; S30: Based on the preprocessed data, extract temporal embedding features using a long short-term memory network model with a time attention mechanism, and combine business features and system monitoring data in the preprocessed data to obtain the probability of order blockage risk using a limit gradient boosting model; S40: Calculate the order system load factor based on system monitoring data, combine the proportion of controlled orders and time-sensitive factors to calculate the impact of regular processes, and determine whether to trigger a controlled short process based on the impact of regular processes and the probability of order congestion risk.

[0014] Among them, Long Short-Term Memory (LSTM) is a time-recurrent neural network, and the eXtreme Gradient Boosting (XGBoost) model is based on gradient boosting decision trees, introducing regularization terms and second-order derivative information to improve performance. Real-time data collected from the vehicle network connectivity management platform is preprocessed to transform it into standardized features suitable for input into the model. Business features are one-hot encoded as package types. The impact of short processes on data closure can be quantified by the influence degree of the regular process, with the value range being [0,1]. The vehicle network connectivity management platform includes a billing system, an order system, and a service activation module. The billing system is used for cost accounting and cost control, the order system is used for scheduling, managing, and recording business processes, and the service activation module is used to issue instructions to network elements for execution. Controlling short processes refers to an emergency handling mechanism where the billing system temporarily establishes a direct communication path with the service activation module, bypassing the standard scheduling queue of the order system, and directly issuing control instructions. The aforementioned vehicle network connectivity management platform control and processing method collects real-time data from the vehicle network connectivity management platform and preprocesses it. Then, it uses a long short-term memory network model with a time attention mechanism combined with a limit gradient boosting model to predict the probability of order congestion risk. This fully utilizes multi-dimensional information and is accurate and reliable, enabling proactive prediction of order congestion risk. This solves the problem of resource waste caused by the backlog and delay in issuing control instructions due to unpredictable order congestion. Based on the probability of order congestion risk and the impact of regular processes, the method determines the triggering of short control processes. This ensures that most regular orders are processed normally while intelligently identifying and triggering short processes for high-priority control orders. This avoids the abuse of short processes and excessive impact on system data consistency, and ensures the priority processing of core control business.

[0015] Combination Figure 2 Preferably, in some embodiments, the preprocessing of the real-time data collected from the vehicle network connection management platform in step S20 includes: S21: Perform data cleaning on the collected real-time data to obtain cleaned data; specifically, the data cleaning of the collected real-time data involves: detecting whether there is missing data in the real-time data, filtering out the missing data in the real-time data, that is, removing the missing data from the real-time data to obtain cleaned data. Missing data may be data that was not collected, or the value corresponding to the missing data may be empty. Therefore, the data cleaning of the collected real-time data involves: removing data with empty values ​​from the collected real-time data to obtain cleaned data.

[0016] S22: Perform outlier correction on the cleaned data to obtain corrected data; wherein, the corrected data refers to the cleaned data after outlier correction; S23: Perform feature standardization on the corrected data to obtain standardized data.

[0017] Specifically, the outlier correction in step S22 includes: The 3σ criterion is used to identify outliers in the dynamic temporal characteristics of the cleaned data; The outliers identified are corrected using the adjacent time-series difference method.

[0018] The dynamic time-series characteristics include order system data and package data, which include real-time order volume, order growth rate, length of pending order message queue, average order processing time, number of packages expiring, number of data overage cards, and number of package renewals.

[0019] Preferably, in some embodiments, the step of identifying outliers using the 3σ criterion for the dynamic temporal characteristics in the cleaned data includes: A sliding window is set for the dynamic temporal characteristics in the cleaned data; wherein, the length of the sliding window is set according to the data sampling frequency; Calculate the mean and standard deviation of the dynamic time-series features of each dimension within the sliding window; The difference between the mean of the dynamic time series features of each dimension and three times the standard deviation of the dynamic time series features of the corresponding dimension is used as the preset minimum threshold of the dynamic time series features of each dimension. The sum of the mean of the dynamic time series features of each dimension and three times the standard deviation of the dynamic time series features of the corresponding dimension is used as the preset maximum threshold of the dynamic time series features of each dimension. Dynamic time series features whose values ​​are less than the preset minimum threshold of the corresponding dimension's dynamic time series features, or whose values ​​are greater than the preset maximum threshold of the corresponding dimension's dynamic time series features, are considered as abnormal dynamic time series features. The values ​​corresponding to the abnormal dynamic time series features are outliers.

[0020] Specifically, the adjacent time-series difference method used to correct outliers obtained through adjacent time-series difference is expressed by formula (1): (1) In the formula, The dynamic temporal characteristics of numerical anomalies are represented by the value of the previous time step in relation to the current time step. The value of the dynamic temporal characteristics of numerical anomalies at the next time step after the current time step. This represents the value of the dynamic time series characteristics at the current moment after correction.

[0021] Specifically, step S23, namely the feature standardization of the corrected data, includes: Based on a preset duration, the key time-series features in the corrected data are used to construct a time-series matrix in chronological order; wherein, the key time-series features include real-time order volume, length of the pending order message queue, order growth rate, average order processing time, and number of package expirations; the preset duration can be 240 minutes; Calculate the historical mean and historical standard deviation for each dimension's key time-series features over a specified period; the specified period can be 3 months. The time series matrix is ​​standardized using the Z-score algorithm to obtain the standardized time series matrix.

[0022] Preferably, the standard score algorithm can be expressed by formula (2): (2) In the formula, Represents the standardized time series matrix. Represents the time series matrix. This represents the historical average of key time-series features over a given period. This represents the historical standard deviation of key time-series characteristics over a given time period.

[0023] Combination Figure 3 Specifically, step S30, which involves extracting temporal embedding features from the preprocessed data using a Long Short-Term Memory network model with a temporal attention mechanism, and combining business features and system monitoring data from the preprocessed data with a limit gradient boosting model to obtain the probability of order congestion risk, includes: S31: Extract temporal embedding features from the standardized temporal matrix using a long short-term memory network with a temporal attention mechanism to obtain the probability of the first block; S32: Construct a fusion feature matrix based on the business characteristics, the time-series embedding features, and the system monitoring data in the corrected data; wherein, the dimension of the fusion feature matrix is... The business features have a dimension of 5, the time-series embedded features can have a dimension of 64, and the system monitoring data has a dimension of 3.

[0024] S33: Input the fused feature matrix into the extreme gradient boosting model to obtain the second single-block probability; S34: Weight and fuse the first and second blockage probabilities to obtain the blockage risk probability.

[0025] Specifically, step S31, which involves extracting temporal embedding features from the standardized temporal matrix using a long short-term memory network with a temporal attention mechanism, includes: Calculate the attention weights of the hidden states at each time step of the Long Short-Term Memory network; The hidden states at each time step are weighted and summed according to their corresponding attention weights to obtain temporal embedding features.

[0026] The structure of the Long Short-Term Memory (LSTM) network includes an input layer, a first LSTM network layer, a second LSTM network layer, and a temporal embedding layer. The first LSTM network layer contains 64 neurons, and the second LSTM network layer contains 32 neurons. The probability of Dropout in the first LSTM network layer is 0.2. Dropout refers to randomly selecting a group of neurons to ignore during the training phase. Ignored units are units that are not considered in a specific forward or backward propagation process.

[0027] Specifically, the attention weights of the hidden state can be calculated using formula (3): (3) In the formula, The first part represents the Long Short-Term Memory network. Attention weights for the hidden states at each time step. and Indicates the sequence number of the time step. and Indicates learnable parameters, and These represent the first and second halves of the Long Short-Term Memory network, respectively. The time step and the first The hidden state at each time step.

[0028] Preferably, the temporal embedding features can be calculated using formula (4): (4) In the formula, Representing temporal embedding features, The first part represents the Long Short-Term Memory network. The hidden state at each time step The first part represents the Long Short-Term Memory network. Attention weights for the hidden states at each time step.

[0029] Specifically, the probability of the first blocked order can be calculated using formula (5): (5) In the formula, This represents the probability of the first block. This represents the Sigmoid function. Representing temporal embedding features, and This represents the learnable parameters.

[0030] Specifically, the probability of order blocking risk can be calculated using formula (6): (6) In the formula, Indicates the probability of order congestion. This represents the probability of the first block. Indicates the fusion weight. This represents the probability of the second blocked order. The value of the fusion weight, determined by minimizing the validation set error, is 0.6, i.e. .

[0031] Specifically, the order system load factor in step S40 can be calculated using formula (7): (7) In the formula, express The order system load factor at any given time. express CPU utilization at any given time express Memory utilization at any given time express Database read / write latency at any given moment, measured in milliseconds.

[0032] Specifically, the influence of the conventional process in step S40 can be calculated using formula (8): (8) In the formula, express The impact of the routine process at any given time. express The order system load factor at any given time. express The proportion of time-sensitive control orders. This represents the time sensitivity factor. The time sensitivity factor is 1 at the end and beginning of the month, and 0.5 at other times. The proportion of time-controlled orders can be determined according to The ratio of the number of controlled orders at any given time to the total number of controlled orders is calculated.

[0033] Specifically, step S40, which involves determining whether to trigger a control-short process based on the impact of the regular process and the probability of order congestion, includes: Compare the probability of order congestion risk with the preset order congestion risk threshold; When the probability of order blockage is not less than the preset order blockage risk threshold, the impact of the normal process is compared with the preset impact threshold. When the impact of the regular process is not greater than the preset impact threshold, the control short process is triggered.

[0034] Preferably, after comparing the congestion risk probability with a preset congestion risk threshold, the vehicle network connection management platform control and processing method further includes: When the probability of order blockage is not less than 0.7, the control short process will not be triggered, and the original process will be executed.

[0035] Specifically, after comparing the impact of the conventional process with the preset impact threshold, the vehicle network connection management platform control and processing method further includes: When the impact of the regular process is greater than the preset impact threshold, the impact of the regular process at the corresponding time after a preset time interval is calculated as the updated impact of the regular process. The updated impact of the regular process is compared with the preset impact threshold. When the updated impact of the regular process is greater than the preset impact threshold, the network outage orders in the controlled orders are processed first.

[0036] Among them, controlled orders include shutdown orders and network outage orders. The preset congestion risk threshold is 0.7, the preset impact threshold is 0.5, and the preset time interval is 10 minutes. When the congestion risk probability is not less than 0.7, the controlled short process is not triggered, and the original process is executed. When the congestion risk probability is not less than 0.7 and the impact of the regular process is not greater than 0.5, the controlled short process is triggered. When the congestion risk probability is not less than 0.7 and the impact of the regular process is greater than 0.5, the impact of the regular process at the corresponding time after a 10-minute delay is calculated as the updated impact of the regular process. The updated impact of the regular process is compared with the preset impact threshold. When the updated impact of the regular process is greater than 0.5, the network outage orders in the controlled orders are processed first.

[0037] Specifically, after determining whether the control short process is triggered in step S40, the vehicle network connection management platform control processing method further includes: When a short control process is determined to be triggered, a control instruction is generated and issued for execution. The corresponding orders are entered and stored according to the control instructions.

[0038] The control commands include shutdown commands and network disconnection commands. When a control short process is triggered, a control command is generated. The billing system simulates the order system sending the control order via an interface protocol to the service activation module. The service activation module then distributes the control command to the network element and executes it. Once the network element has completed execution, the service activation module sends a completion signal back to the billing system. The billing system then sends an order supplementation process to the order system to supplement the corresponding order information according to the control command and stores the supplemented order information.

[0039] As can be seen, in the above solution, by collecting and preprocessing real-time data from the vehicle network connection management platform, a long short-term memory network model with a time attention mechanism is used in conjunction with a limit gradient boosting model to predict the probability of order congestion. This fully utilizes multi-dimensional information and is accurate and reliable, enabling proactive prediction of order congestion risks. This solves the problem of resource waste caused by the backlog and delay in issuing control instructions due to unpredictable order congestion. Based on the probability of order congestion risk and the impact of regular processes, the solution makes judgments on triggering short control processes. This ensures that most regular orders are processed normally, while intelligently identifying and triggering short processes for high-priority control orders. This avoids the abuse of short processes and excessive impact on system data consistency, and ensures the priority processing of core control business.

[0040] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0041] In one embodiment, a vehicle network connectivity management platform control and processing device is provided, which corresponds one-to-one with the vehicle network connectivity management platform control and processing method described in the above embodiments. For example... Figure 4 As shown, the vehicle network connectivity management platform control and processing device includes a data acquisition module 101, a data preprocessing module 102, a fusion prediction module 103, and a dynamic decision-making module 104. Detailed descriptions of each functional module are as follows: The data acquisition module 101 is used to collect real-time data from the vehicle network connection management platform, wherein the real-time data includes order system data, system monitoring data, and package data; The data preprocessing module 102 is used to preprocess the real-time data collected from the vehicle network connection management platform to obtain preprocessed data from the vehicle network connection management platform. The fusion prediction module 103 is used to extract temporal embedding features from the preprocessed data through a long short-term memory network model with a time attention mechanism, and combine business features and system monitoring data in the preprocessed data through a limit gradient boosting model to obtain the probability of order blockage risk. The dynamic decision-making module 104 is used to calculate the order system load factor based on system monitoring data, calculate the impact of the regular process by combining the proportion of controlled orders and the time-sensitive factor, and determine whether to trigger the controlled short process based on the impact of the regular process and the probability of order blockage risk.

[0042] In one embodiment, the data preprocessing module 102 is specifically used for: The collected real-time data is cleaned to obtain cleaned data; The cleaning data is corrected for outliers to obtain corrected data; The corrected data is then subjected to feature standardization to obtain standardized data.

[0043] In one embodiment, the data preprocessing module 102 is specifically used for: The 3σ criterion is used to identify outliers in the dynamic temporal characteristics of the cleaned data; The outliers identified are corrected using the adjacent time-series difference method.

[0044] In one embodiment, the data preprocessing module 102 is further configured to: Based on a preset duration, the key time-series features in the corrected data are used to construct a time-series matrix in chronological order; wherein, the key time-series features include real-time order volume, length of the pending order message queue, order growth rate, average order processing time, and number of package expirations; Calculate the historical mean and historical standard deviation of the key time-series features for each dimension at a set time. The time series matrix is ​​standardized using the standard score algorithm to obtain the standardized time series matrix.

[0045] In one embodiment, the fusion prediction module 103 is specifically used for: The standardized temporal matrix is ​​used to extract temporal embedding features through a long short-term memory network with a time attention mechanism to obtain the probability of the first block. A fusion feature matrix is ​​constructed based on business characteristics, the time-series embedded features, and the system monitoring data in the corrected data; The fused feature matrix is ​​input into the extreme gradient boosting model to obtain the second single-block probability. The probability of the first blockage and the probability of the second blockage are weighted and fused to obtain the probability of the blockage risk.

[0046] In one embodiment, the dynamic decision-making module 104 is specifically used for: Compare the probability of order congestion risk with the preset order congestion risk threshold; When the probability of order blockage is not less than the preset order blockage risk threshold, the impact of the normal process is compared with the preset impact threshold. When the impact of the regular process is not greater than the preset impact threshold, the control short process is triggered.

[0047] In one embodiment, the dynamic decision-making module 104 is further configured to: When a short control process is determined to be triggered, a control instruction is generated and issued for execution. The corresponding orders should be entered in accordance with the control instructions.

[0048] This invention provides a control and processing device for a vehicle network connectivity management platform. By collecting and preprocessing real-time data from the vehicle network connectivity management platform, it uses a long short-term memory network model with a time attention mechanism combined with a limit gradient boosting model to predict the probability of order congestion. This fully utilizes multi-dimensional information and is accurate and reliable, enabling proactive prediction of order congestion risks. It solves the problem of resource waste caused by unpredictable order congestion leading to delayed issuance of control instructions. Based on the probability of order congestion risk and the impact of regular processes, it determines the triggering of short control processes. This ensures that most regular orders are processed normally while intelligently identifying and triggering short processes for high-priority control orders, avoiding excessive impact on system data consistency due to the abuse of short processes, and ensuring the priority processing of core control business.

[0049] Specific limitations regarding the vehicle-to-everything (V2X) connectivity management platform control and processing device can be found in the limitations of the V2X connectivity management platform control and processing method described above, and will not be repeated here. Each module in the aforementioned V2X connectivity management platform control and processing device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the corresponding operations of each module.

[0050] In one embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 5As shown, the computer device includes a processor, memory, network interface, and database connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile and / or volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The network interface is used to communicate with external clients via a network connection. When the computer program is executed by the processor, it implements the functions or steps of a vehicle network connection management platform control processing method on the server side.

[0051] In one embodiment, a computer device is provided, which may be a client, and its internal structure diagram may be as follows: Figure 6 As shown, the computer device includes a processor, memory, network interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The network interface is used to communicate with an external server via a network connection. When the computer program is executed by the processor, it implements the functions or steps of a vehicle network connection management platform control processing method on the client side.

[0052] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to perform the following steps: Collect real-time data from the vehicle network connection management platform, including order system data, system monitoring data, and package data; The real-time data collected from the vehicle network connection management platform is preprocessed to obtain preprocessed data from the vehicle network connection management platform. Based on the preprocessed data, temporal embedding features are extracted using a long short-term memory network model with a time attention mechanism. Combined with business features and system monitoring data in the preprocessed data, the probability of order blockage risk is obtained using a limit gradient boosting model. The order system load factor is calculated based on system monitoring data. The impact of the regular process is calculated by combining the proportion of controlled orders and the time-sensitive factor. Based on the impact of the regular process and the probability of order congestion, it is determined whether to trigger the controlled short process.

[0053] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, the computer program performing the following steps when executed by a processor: Collect real-time data from the vehicle network connection management platform, including order system data, system monitoring data, and package data; The real-time data collected from the vehicle network connection management platform is preprocessed to obtain preprocessed data from the vehicle network connection management platform. Based on the preprocessed data, temporal embedding features are extracted using a long short-term memory network model with a time attention mechanism. Combined with business features and system monitoring data in the preprocessed data, the probability of order blockage risk is obtained using a limit gradient boosting model. The order system load factor is calculated based on system monitoring data. The impact of the regular process is calculated by combining the proportion of controlled orders and the time-sensitive factor. Based on the impact of the regular process and the probability of order congestion, it is determined whether to trigger the controlled short process.

[0054] It should be noted that the functions or steps that can be implemented by the computer-readable storage medium or computer device described above can be referred to the relevant descriptions on the server side and client side in the foregoing method embodiments. To avoid repetition, they will not be described one by one here.

[0055] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0056] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is used as an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0057] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A method for managing and controlling a vehicle-to-everything (V2X) connectivity management platform, characterized in that: Includes the following steps: Collect real-time data from the vehicle network connection management platform, including order system data, system monitoring data, and package data; The real-time data collected from the vehicle network connection management platform is preprocessed to obtain preprocessed data from the vehicle network connection management platform. Based on the preprocessed data, temporal embedding features are extracted using a long short-term memory network model with a time attention mechanism. Combined with business features and system monitoring data in the preprocessed data, the probability of order blockage risk is obtained using a limit gradient boosting model. The order system load factor is calculated based on system monitoring data. The impact of the regular process is calculated by combining the proportion of controlled orders and the time-sensitive factor. Based on the impact of the regular process and the probability of order congestion, it is determined whether to trigger the controlled short process.

2. The vehicle network connection management platform control and processing method according to claim 1, characterized in that, The preprocessing of the real-time data collected from the vehicle network connection management platform includes: The collected real-time data is cleaned to obtain cleaned data; The cleaning data is corrected for outliers to obtain corrected data; The corrected data is then subjected to feature standardization to obtain standardized data.

3. The vehicle network connection management platform control and processing method according to claim 2, characterized in that, The outlier correction of the cleaned data includes: The 3σ criterion is used to identify outliers in the dynamic temporal characteristics of the cleaned data; The outliers identified are corrected using the adjacent time-series difference method.

4. The vehicle network connection management platform control and processing method according to claim 2, characterized in that, The feature standardization of the corrected data includes: Based on a preset duration, the key time-series features in the corrected data are used to construct a time-series matrix in chronological order; wherein, the key time-series features include real-time order volume, length of the pending order message queue, order growth rate, average order processing time, and number of package expirations; Calculate the historical mean and historical standard deviation of the key time-series features for each dimension at a set time. The time series matrix is ​​standardized using the standard score algorithm to obtain the standardized time series matrix.

5. The vehicle network connection management platform control and processing method according to claim 4, characterized in that, The step of extracting temporal embedding features from the preprocessed data using a Long Short-Term Memory network model with a temporal attention mechanism, and combining business features and system monitoring data from the preprocessed data with a Limiting Gradient Boosting model to obtain the probability of order congestion risk, includes: The standardized temporal matrix is ​​used to extract temporal embedding features through a long short-term memory network with a time attention mechanism to obtain the probability of the first block. A fusion feature matrix is ​​constructed based on business characteristics, the time-series embedded features, and the system monitoring data in the corrected data; The fused feature matrix is ​​input into the extreme gradient boosting model to obtain the second single-block probability. The probability of the first blockage and the probability of the second blockage are weighted and fused to obtain the probability of the blockage risk.

6. The vehicle network connection management platform control and processing method according to claim 1, characterized in that, The step of determining whether to trigger a controlled short process based on the impact of the regular process and the probability of order congestion includes: Compare the probability of order congestion risk with the preset order congestion risk threshold; When the probability of order blockage is not less than the preset order blockage risk threshold, the impact of the normal process is compared with the preset impact threshold. When the impact of the regular process is not greater than the preset impact threshold, the control short process is triggered.

7. The vehicle network connection management platform control and processing method according to claim 1, characterized in that, After determining whether a short control process has been triggered, the vehicle network connection management platform control and processing method further includes: When a short control process is determined to be triggered, a control instruction is generated and issued for execution. The corresponding orders should be entered in accordance with the control instructions.

8. A vehicle network connection management platform control and processing device, characterized in that, include: The data acquisition module is used to collect real-time data from the vehicle network connection management platform, wherein the real-time data includes order system data, system monitoring data, and package data; The data preprocessing module is used to preprocess the real-time data collected from the vehicle network connection management platform to obtain preprocessed data from the vehicle network connection management platform. The fusion prediction module is used to extract temporal embedding features from the preprocessed data using a long short-term memory network model with a time attention mechanism, and to obtain the probability of order congestion risk by combining business features and system monitoring data in the preprocessed data with a limit gradient boosting model. The dynamic decision-making module is used to calculate the order system load factor based on system monitoring data, combine the proportion of controlled orders and time-sensitive factors to calculate the impact of regular processes, and determine whether to trigger a controlled short process based on the impact of regular processes and the probability of order congestion risk.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the vehicle network connection management platform control and processing method as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the vehicle network connection management platform control and processing method as described in any one of claims 1 to 7.

Images